High concordance of PET-CT treatment response evaluation according to PERCIST 1.0 when comparing images reconstructed with OSEM vs. BSREM.

Background: Positron emission tomography (PET) response criteria in solid tumours (PERCIST 1.0) is a systematic assessment of therapy response with fluorine 18 fluorodeoxyglucose ([F]FDG) PET. The ordered subset expectation maximization (OSEM) and block sequential regularized expectation maximization (BSREM) reconstruction algorithms are both commonly used. The impact of using OSEM or BSREM on PERCIST 1.0 is not fully determined. This study aimed to compare treatment response evaluation between the reconstructions OSEM and BSREM according to PERCIST 1.0.

Methods: The highest peak standardized uptake value corrected for lean body mass (SUL) in a lesion were measured in PET-CT scans from 90 patients with metastatic breast cancer or malignant melanoma, who had undergone two [F]FDG PET with computer tomography (CT) scans for treatment evaluation purpose. SUL measurements in the reference organ were also calculated. All scans were reconstructed with OSEM and BSREM. Bland Altman plots and Spearman correlation coefficient were used for comparing measurements obtained from OSEM and BSREM images. Patients were categorized as a complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), or progressive metabolic disease (PMD).

Results: The mean standard deviation (SD) of SUL in the reference organ was slightly lower for BSREM images, resulting in a slightly lower threshold. Lesion SUL was slightly higher for BSREM images compared with OSEM images. Only 5 out of 90 patients had different response evaluations for the different reconstructions.

Conclusion: PERCIST 1.0 serves as a robust response evaluation criterion and is only marginally dependent on the reconstruction used.